ELMER v.2: An R/Bioconductor package to reconstruct gene regulatory networks from DNA methylation and transcriptome profiles

Abstract

Recent studies indicate that DNA methylation can be used to identify changes at transcriptional enhancers and other cis-regulatory elements in primary human samples. A systematic approach to inferring gene regulatory networks has been provided by the R/Bioconductor package ELMER (Enhancer Linking by Methylation/Expression Relationships), which first identifies DNA methylation changes in distal regulatory elements and correlates these with the expression of nearby genes to identify direct transcriptional targets. Next, ELMER performs a transcription factor binding motif analysis and integrates with expression profiling of all human transcription factors, to identify master regulatory TFs and place each differentially methylated regulatory element into the context of an altered gene regulatory network (GRN).Here we present a completely updated version of the package (ELMER v. 2.0), which uses the latest Bioconductor data structures including the popular MultiAsssayExperiment, supports multiple reference genome assemblies as well as the DNA methylation platforms Infinium MethylationEPIC and Infinium HumanMethylation450, and provides a "Supervised" analysis mode for paired sample study designs (such as treated vs. untreated replicate samples). It also supports data import from the new NCI Genomic Data Commons (GDC) database. The new version is substantially re-written, improving stability, performance, and extensibility. It also uses improved databases for transcription factor binding domain families and binding motif specificities, and has newly designed output plots for publication-quality figures.Below, we describe the methods and new features of ELMER v. 2.0 and present two use case demonstrating how the tool can be used to analyze TCGA data in either Unsupervised or Supervised mode. ELMER (v2.0.0) is available as an R/Bioconductor package at

Lineage-Specific Epigenomic and Genomic Activation of the Oncogene HNF4A Promotes Gastrointestinal Adenocarcinomas

“…Illumina HM450 methylation data from 4 different TCGA projects, COAD-READ, ESCA, PAAD, and STAD were processed with SeSAMe[51], and the matched RNA-seq (FPKM-UQ) data were downloaded from GDC (Genomic Data Commons, https://portal.gdc.cancer.gov/) from the harmonized database (data aligned to hg38/GRCh38). Unsupervised analysis was performed using ELMER[52] for the following pair-wise comparisons: EAC primary tumors (n = 78) vs ESCC primary tumors (n = 76); EAC primary tumors (n = 78) vs EAC normal adjacent samples (n = 5), COAD-READ primary tumors (n = 389) vs colon normal adjacent samples (n = 21), PAAD primary tumors (n = 177) vs pancreas normal adjacent samples (n = 4). Because of the small number of normal adjacent samples from STAD (n=2) which also lacked RNA-seq data, we could not perform comparison for STAD primary tumors.…”

“…Our earlier work demonstrated that this form of de-methylation could be used to identify cancer-specific enhancers which were strongly enriched for specific transcription factor binding sites (TFBSs)[14]. Recently, we have developed a novel computational algorithm, ELMER (Enhancer Linking by Methylation/Expression Relationships), to identify and exploit systematically these TFBS-associated methylation changes in cancer[17, 18]. Briefly, the approach of ELMER is to utilize TFBS methylation changes as key nodes within the larger gene regulatory network, and correlate with gene expression to infer both the upstream (master regulator TFs, MRTFs) and downstream (target genes) links for each TFBS.…”

“…Specifically, when a group of enhancers is coordinately altered in a specific sample subset, this is often the result of an altered upstream MRTF in the gene regulatory network. To identify MRTFs involved in setting the tumor-specific DNA methylation patterns, ELMER correlates DNA methylation occurring within binding sites for specific MRTFs (inferred by specific TF binding motifs), with altered expression of the corresponding MRTF[17, 18].…”

TCGAbiolinksGUI: A graphical user interface to analyze GDC cancer molecular and clinical data

“…Also, using the DNA methylation data and the gene expression data the R/Bioconductor ELMER package can be used to discover functionally relevant genomic regions associated with cancer [7,8].…”

“…The following R/Bioconductor packages are used as back-ends for the data retrieval and analysis: TCGAbiolinks [2] which allows to search, download and prepare data from the NCI's Genomic Data Commons (GDC) data portal into an R object and perform several downstream analysis; ELMER (Enhancer Linking by Methylation/Expression Relationship) [7,8] which identifies DNA methylation changes in distal regulatory regions and correlate these signatures with the expression of nearby genes to identify transcriptional targets associated with cancer; ComplexHeatmap [9] to visualize data as oncoprint and heatmaps, pathview [10] which offers pathway based data integration and visualization; and maftools [11] to analyze, visualize and summarize MAF (Mutation Annotation Format) files.…”